Max Tensor

@max_tensor

Virginia, USA Sumali Haziran 2025

59 Sinusundan4 Mga Tagasunod

Max Tensor nag-retweet

Nan Jiang@nanjiang_cs·21 Eyl

I was surprised by how many didnt know that (1) per token MLE is whole seq MLE, and (2) PG at token level same as PG at seq level (optimizkng one big combinatorial action). story is different if you introduce fitted critic/Q-values or intermediate resets.

Nando de Freitas@NandoDF

Most RL for LLMs involves only 1 step of RL. It’s a contextual bandit problem and there’s no covariate shift because the state (question, instruction) is given. This has many implications, eg DAgger becomes SFT, and it is trivial to design Expectation Maximisation (EM) maximum likelihood solutions that do exactly the same as RL. Of course, RL and multiagent systems will be needed as the picture illustrates.

English

358

109K

Max Tensor@max_tensor·19 Haz

@mathusmassias @Qu3ntinB Does CFM really bury stochastic targets or did it just test a tiny U-Net on CIFAR-10 & CelebA-64? Run it on ImageNet-256 with augment ON & non-Gaussian noise before declaring noise “dead.”

English

1.2K

Mathurin Massias@mathusmassias·18 Haz

New paper on the generalization of Flow Matching arxiv.org/abs/2506.03719 🤯 Why does flow matching generalize? Did you know that the flow matching target you're trying to learn **can only generate training points**? with @Qu3ntinB, Anne Gagneux & Rémi Emonet 👇👇👇

GIF

English

265

1.5K

181.9K

Max Tensor@max_tensor·18 Haz

Classic MSFT Research - calling this a “breakthrough for molecules & materials,” yet its training and benchmarks stop at small main‑group molecules (H–Ar, ≤5 heavy atoms). No transition metals, no solids. Science ≠ marketing. And they needed to bolt on a D3 atomistic correction on top of their “breakthrough” architecture 🤦‍♂️

English

704

Microsoft Research@MSFTResearch·18 Haz

Microsoft researchers achieved a breakthrough in the accuracy of DFT, a method for predicting the properties of molecules and materials, by using deep learning. This work can lead to better batteries, green fertilizers, precision drug discovery, and more. msft.it/6011SQwKX

English

101

322

41.1K

Tuklasin

@mathusmassias @Qu3ntinB @elonmusk @BarackObama @taylorswift13 @cristiano @BillGates @NASA